Intra-amygdala anxiogenic drug infusion prior to retrieval biases rats towards the use of habit memory

In a dual-solution plus-maze task that can be acquired using either hippocampus-dependent “cognitive/place” learning or dorsal striatal-dependent “habit/response” learning, pre-acquisition peripheral or intra-basolateral amygdala (BLA) injections of anxiogenic drugs result in the predominant use of response learning. The present experiments examined the effect of anxiogenic drug treatment on the relative use of multiple memory systems when administered prior to memory retrieval. Adult male Long-Evans rats were trained for two days (6 trials/day, 30 s ITI) in a dual-solution plus-maze task to swim from the same start point (south) to an escape platform that was located in a consistent goal arm (west). On day three, prior to a memory retrieval probe trial from a novel start point (north), rats received a peripheral (0.03, 0.1 or 0.3 mg/kg), or intra-BLA (0.1 μg/0.5 μl) injection of the anxiogenic α2-adrenoreceptor antagonist RS 79948-197, or saline. Relative to saline controls, rats receiving either peripheral or intra-BLA infusions of RS 79948-197 predominantly displayed response learning on the probe trial. In an additional experiment peripheral (0.1 mg/kg) or intra-BLA (0.1 μg) drug injections administered prior to both acquisition and retrieval also resulted in the predominant use of response learning. The findings indicate that (1) similar to acquisition, peripheral injection of an anxiogenic drug prior to memory retrieval biases rats towards the use of habit/response memory, (2) intra-BLA infusions of an anxiogenic drug is sufficient to produce this modulatory effect of emotional state on memory retrieval, and (3) state-dependency does not appear to play a role in the effects of anxiogenic drug treatment on multiple memory system use. The findings may have implications for understanding the interaction between brain function, emotion, and the relative use of multiple memory systems in human psychopathology.